Appendix 2. Methods: Systematic review

The study by Wollscheid (See chapter 2) applies a rapid systematic review. A systematic review is a form of secondary research that applies scientific methods and includes existing studies as data material. The approach is less resource-intensive than a full systematic review, but it still adheres to scientific methods. Less resource-intensive systematic reviews have been developed in response to the need for reliable evidence as a basis for political decision-making. The literature search (data collection) is limited in scope and is restricted by, for example, time frame, databases, and/or language (Thomas et al., 2013; Wollscheid & Tripney, 2021). Figure 1 provides an overview of different stages in systematic reviews.

Figure 4. Overview of different stages in systematic reviews

1. Operationalisation of review questions

2. Search strategy and literature search

3. Study selection

4. Mapping analyses and synthesis

Eligibility criteria

The review includes research about second-language learning and teaching of students with foreign-language background in Nordic countries. The main question to be addressed in the literature review is:

What programmes are being used in the Nordic countries to facilitate and support second-language acquisition and instruction in compulsory and upper secondary education? How do these programmes work, and for which groups of students? What is the role of the first language for second-language learning?

Empirical studies addressing national programmes to support second-language acquisition among students with foreign-language background were included, either at compulsory level or upper secondary school level. Studies of programmes that aim at first-language learning or individual pedagogics only are not eligible. To be included, empirical studies must have been published in scientific journals or as a research report between 2014 and 2024. The publication language must be English, Norwegian, Swedish, or Danish. For publications in Finnish and Icelandic, English summaries were used. Among eligible studies addressing second-language learning, specific emphasis was placed on those with an experimental or quasi-experimental design, i.e., studies with a type of control group that might give an indication of how these programmes work, i.e., their effect. Examples include randomised controlled trials and difference-in-difference. As only a few studies with an experimental or quasi-experimental design were expected (see Wollscheid et al., 2017), observational studies and qualitative studies were also included, if they addressed second-language learning. The population–concept–context framework recommended by the Joanna Briggs Institute for scoping reviews was used to operationalise the eligible criteria (Peters et al., 2015).

Table 2. Eligible criteria for the systematic literature review

	Inclusion criteria	Exclusion criteria
P-population (Who?)	Students with foreign-language background	Students with native-language/majority background
C-concept (What?)	National programmes to support language acquisition First-language programmes to facilitate second-language learning Introductory offers Bilingual programmes (including the first language) Pedagogics when intertwined with national language policies	Programmes to support language acquisition in first language Special needs education Programmes supporting second-language acquisition of preschoolers Individual pedagogics, e.g., digital tools for L2 learning
C-context (With what qualifiers?)	Compulsory education Upper secondary education	Higher education Pre-School
Outcome	Main outcome: Second-language learning Secondary outcomes: Other educational outcomes Social outcomes	Teacher outcomes only First-language learning only
Time span	2014–2024	Before 2014
Publication status	Peer–reviewed journal articles, Research reports	Other types of scholarly publications (conference papers)
Language	English, Danish, Swedish, Norwegian, Finnish, Icelandic (with English abstract)	Other languages than English and Nordic
Study design	Empirical studies with control conditions: experimental and quasi-experimental studies, e.g., randomised controlled trials, difference-in-difference, longitudinal studies Evaluations including qualitative and quantitative data Other studies addressing L2	Theoretical papers Small case studies, ethnographic studies, qualitative studies which do not address second-language learning
Geography	Nordic countries	Non-Nordic countries

Search strategy and sources

Research in education is interdisciplinary with heterogeneous study designs and a diverse publication culture comprising national languages in addition to English and different publication channels (Heck et al., 2024). Therefore, systematic searches in several international indexed databases and searches in national databases were combined. At a later stage, different types of supplementary strategic searches were added (e.g., consultation of country experts, snowballing), a common strategy applied in systematic reviews within the social sciences (Papaioannou et al., 2010). Peer-reviewed journal articles and research reports were included to reduce publication bias and included studies with different designs in the first step.

The following databases were searched for the period 2014–2024:

Web of Science Core (WoS) comprises the world’s leading scholarly journals, books, and proceedings in the sciences, social sciences, and arts and humanities and navigates the full citation network. WoS enables limiting the literature according to geography (European countries) and languages (English and Scandinavian languages). WoS has a relatively low coverage of publications in the social sciences and the humanities in particular, and a low coverage of literature published in languages other than English, and book publications (Aksnes & Sivertsen, 2019). Additional searches were therefore conducted in national and field-specific databases, and strategic searches to address these limitations were added.

ERIC (Education Resource Information Center), sponsored by the Institute of Education Sciences (IES) of the U.S. Department of Education, is a leading full-text research database of full-text education journals. It provides scholarly research and information to meet the needs of education students, researchers, and policymakers. It covers all levels of education from early childhood to higher education.

Kudos is a collection of documents such as public reports and white papers. Documents include evaluations, annual reports, strategy documents, and other policy documents. The Kudos database was searched for research reports.

Swepub collects metadata of research publications from Swedish higher education institutions in the form of scientific articles, books, theses, conference papers, etc.

DiVA is a common search function for all publications that have been registered and published with each DiVA member. Each title in the database contains bibliographic information, usually an abstract, and in several cases a link to the full text where the work is published.

IRIS is the Icelandic Research Information System and shows the research activities of Icelandic institutions. The National and University Library of Iceland is responsible for operating and supervising the system on behalf of the Ministry of Education and Culture. IRIS provides a public platform to examine research activity across institutions and the social distribution of research knowledge created in Iceland. The portal makes it possible to view contributions and collaborations of researchers, institutions, and disciplines on a domestic and international level.

FINEEC (Finnish Education Evaluation Centre) is an independent agency responsible for the evaluation of education in Finland. It operates as a separate unit within the Finnish National Agency for Education and carries out evaluations related to education from early childhood education to higher education. The agency acts independently, and its evaluations are free from third-party influence. It has a member and an affiliate database.

FINNA is a search service that collects material from hundreds of Finnish organisations under one roof. Using versatile search functions, one can easily access millions of items, free of charge.

Supplementary strategic searches included consulting country experts, i.e., members of the reference group.

Analyses

The included studies were coded by reference, geography, population, programme, outcomes study design, and main findings. During the first round, a broader sample of studies addressing language learning more broadly was included, which was then reduced to studies with a focus on second-language learning. Further, the included sample of 16 studies was divided according to study design. Studies with an experimental and quasi-experimental approach were prioritised in the analysis.

Limitations

Generally, the scope of this literature review was broad, addressing students with a foreign-language background in compulsory school and upper secondary school and different measures facilitating second-language learning in the Nordic countries with different educational systems. As the researchers did not have high expectations of finding studies on the effect of second-language programmes (see also Wollscheid et al., 2017), the scope was not limited to studies with experimental and quasi-experimental design that are deemed to be most adequate to measure effects. The approach was informed by the systematic review methodology, which requires systematic searches in combination with strategic searches to identify eligible studies and rigor and transparency in the selection and analysis of studies. To achieve a balanced sample and prevent bias in the sample of studies, searches in general databases were combined with searches in specific databases and strategic searches via experts.

Each study, however, has its limitations. It is therefore possible that the researchers have missed eligible studies which were neither indexed in the selected databases nor published on eligible websites, nor identified by strategic searches. This may have introduced a selection bias.